The recent emergence of diffusion tensor imaging (DTI) provides a new contrast mechanism through water diffusion characteristics to investigate the white matter development and integrity in the human brain, and their impact on neuronal functions. Given its unique sensitivity, one of the most valuable and fitting applications of DTI i the investigation of developing brains in children, as many debilitating diseases in children are originated from white matter abnormalities and injuries. For example, Cerebral Palsy (CP), which results from in utero or perinatal injuries to motor pathways, is the most prevalent motor disorder of childhood, affecting 2 to 3 out of every 1,000 live births. Its well-defined white mattr pathology would benefit from the exclusive white matter characterization provided by DTI. However, the current DTI practice lacks sufficient spatial resolvability and subsequent quantitative consistency to characterize the impairment of the complex motor pathway in CP and its recovery during treatment. There are many contributing factors to this inadequacy, chief among them are the technical limitations in achieving high spatial resolution, the systematic distortions inherent in the DTI pulse sequences that introduce distortions and errors in tensor estimation, and the heightened sensitivity to global and local motions due to large diffusion weighting gradients that further complicate the practical utility in patient populations. These drawbacks are exacerbated especially in vulnerable populations such as children. In this proposal, we aim to address these current limitations by developing innovative acquisition solutions to achieve the much needed spatial resolution and fidelity, and to subsequently apply our innovative DTI acquisition methodology to better characterize the brain connectivity in children with CP, leveraging our strong partnership with a long-standing and successful clinical program using an innovative and promising stem cell therapy through umbilical cord blood (UCB) infusion. Specifically, we propose to: 1) achieve high spatial resolution necessary to better delineate complex white matter fiber structure, 2) achieve high spatial fidelity to improve tensor estimation and co-registration with neuroanatomy, 3) achieve greatly reduced motion sensitivity to improve practical utility in pediatric patient populations, 4) acquire and develop pediatric brain connectivity maps in CP to investigate the impairment and recovery of motor pathway and functions in children during UCB stem cell therapy. We anticipate that our innovative acquisition methodology will greatly increase the spatial resolvability and quantitative consistency of DTI measures, which can improve our understanding on the promising effects of stem cell therapy and help design the best treatment plan for children with CP. It is also likely that our new methodology will find broader application in basic and clinical neurosciences at large.